llama.cpp
llama : pad KV cache size
#4280
Merged
Overview
Commits
2
Changes
View On GitHub

llama : pad KV cache size #4280

ggerganov merged 2 commits into master from gg/pad-kv-cache
ggerganov
slaren
ggerganov llama : pad KV cache size to 32
3e68df86
ggerganov ggerganov force pushed to 3e68df86 2 years ago
ggerganov metal : try to improve batched decoding
3cb1c348
ggerganov ggerganov force pushed to 3cb1c348 2 years ago
ggerganov
jhen0409
slaren
ggerganov ggerganov merged d7b800b8 into master 2 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum